### What this PR does / why we need it?
This addresses the issue brought up by #5356 and #4963, and we believe
the unnecessary conditions are the root cause.
Change the unpad trigger to be driven by actual size mismatches
(num_reqs vs base_num_reqs or scheduled vs input token counts) rather
than specific speculative-method flags. Then remove brittle workarounds
that forced request counts and sliced query start locations.
This prevents incorrect indexing and length mismatches during
speculative decoding and makes metadata unpadding more robust across
scheduling modes.
### Does this PR introduce _any_ user-facing change?
None.
### How was this patch tested?
Tested by existing cases.
- vLLM version: v0.13.0
- vLLM main:
8be6432bda
---------
Signed-off-by: Yizhou Liu <liu_yizhou@outlook.com>
### What this PR does / why we need it?
Add Qwen3-Omni-30B-A3B-Thinking Tutorials
### Does this PR introduce _any_ user-facing change?
No
### How was this patch tested?
- vLLM version: v0.13.0
- vLLM main:
5326c89803
---------
Signed-off-by: Meihan-chen <jcccx.cmh@gmail.com>
### What this PR does / why we need it?
#5230 this PR introduced a problem when both mtp and full_decode_only
are enabled for the DSV32 model, the operators cannot be compiled into
the graph. This PR fixes that issue.
- vLLM version: v0.13.0
- vLLM main:
2f4e6548ef
Signed-off-by: cookieyyds <126683903+cookieyyds@users.noreply.github.com>
### What this PR does / why we need it?
Adapted sp to eagle3.
There may still be some problems, e.g., accuracy in some scenes,
`sp`+`dp`...
We will fix them later.
### Does this PR introduce _any_ user-facing change?
N/A
### How was this patch tested?
We tested it mainly in a new `e2e`.
```shell
pytest -s tests/e2e/singlecard/spec_decode/test_v1_spec_decode.py::test_llama_qwen_eagle_acceptance
```
```text
.
=============================== warnings summary ===============================
<frozen importlib._bootstrap>:241
<frozen importlib._bootstrap>:241: DeprecationWarning: builtin type SwigPyPacked has no __module__ attribute
<frozen importlib._bootstrap>:241
<frozen importlib._bootstrap>:241: DeprecationWarning: builtin type SwigPyObject has no __module__ attribute
-- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html
============= 3 passed, 1 skipped, 2 warnings in 142.05s (0:02:22) =============
```
It passed.
- vLLM version: v0.13.0
- vLLM main:
7157596103
Signed-off-by: drslark <slarksblood@qq.com>
### What this PR does / why we need it?
Correcting some outdated use cases:
`tests/e2e/singlecard/test_aclgraph_accuracy.py::test_models_output` ->
`tests/e2e/singlecard/test_aclgraph_accuracy.py::test_piecewise_res_consistency`
### Does this PR introduce _any_ user-facing change?
### How was this patch tested?
- vLLM version: v0.13.0
- vLLM main:
2f4e6548ef
Signed-off-by: wangli <wangli858794774@gmail.com>
### What this PR does / why we need it?
EPLB currently does not have CI related to aclgraph and redundancy
experts; this PR adds them.
release on #5529
### Does this PR introduce _any_ user-facing change?
### How was this patch tested?
Tested the use cases to be added in this PR.
PASSED
====================================================== warnings summary
==========================================================
<frozen importlib._bootstrap>:241
<frozen importlib._bootstrap>:241: DeprecationWarning: builtin type
SwigPyPacked has no __module__ attribute
<frozen importlib._bootstrap>:241
<frozen importlib._bootstrap>:241: DeprecationWarning: builtin type
SwigPyObject has no __module__ attribute
-- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html
====================================================== 1 passed, 2
warnings in 272.24s (0:04:32)
=====================================================
- vLLM version: v0.13.0
- vLLM main:
8be6432bda
Signed-off-by: shenchuxiaofugui <1311027364@qq.com>
ref doesn't work with workflow_dispatch, let's change it to raw way
This PR also merge the pr_create job into one runner to save resource.
- vLLM version: v0.13.0
- vLLM main:
2f4e6548ef
Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
### What this PR does / why we need it?
Optimize the warning print information format when detects depredated
code is used in vllm-ascend.
### Does this PR introduce _any_ user-facing change?
NA
- vLLM version: v0.13.0
- vLLM main:
2f4e6548ef
---------
Signed-off-by: leo-pony <nengjunma@outlook.com>
### What this PR does / why we need it?
1. MagicMTP (paper: "Block Verification Accelerates Speculative
Decoding") was introduced to consider the influence among multiple draft
tokens, improving the acceptance rate without compromising accuracy.
2. Added Triton and PyTorch implementations, and added E2E test cases.
### Does this PR introduce _any_ user-facing change?
MagicMTP will automatically take effect when the parameter
"num_speculative_tokens" >= 3.
- vLLM version: v0.13.0
- vLLM main:
7157596103
Signed-off-by: chenaoxuan <cax1165@163.com>
### What this PR does / why we need it?
func `get_kv_cache_spec` in model_runner changed a lot and caused error
in cpuoffloading connector which is copied from model_runner, this PR
adapts to new implemented `get_kv_cache_spec` to fix it.
### How was this patch tested?
- vLLM version: v0.11.0
- vLLM main:
2918c1b49c
Signed-off-by: lidenghui <lidenghui1110@gmail.com>
### What this PR does / why we need it?
- Delete the environment variable
`VLLM_ASCEND_ENABLE_FLASHCOMM2_OSHARED`
- Introduce layer_sharding as a configurable feature in
additional_config
- Revise the term "shared weight" to "shard weight."
Configuration : The feature is opt-in via the additional_config
argument:
```
--additional-config '{
"layer_sharding": ["o_proj", "q_b_proj"]
}'
```
This is orthogonal to standard tensor parallelism and weight replication
strategies. It is treated as a separate, explicit feature.It can be used
in any scenario, combined with the
flashcomm2https://github.com/vllm-project/vllm-ascend/pull/3232 feature
or the ShardedCP #4702 feature, to achieve significant performance.
- vLLM version: v0.12.0
- vLLM main:
ad32e3e19c
---------
Signed-off-by: zzhx1 <zzh_201018@outlook.com>
Signed-off-by: zzhxx <zhangzihang23@mails.ucas.ac.cn>
Signed-off-by: chenxiao <Jaychou1620@Gmail.com>
Co-authored-by: clrs97 <524936896@qq.com>
Co-authored-by: Levi-JQ <yujinqi2@huawei.com>
Co-authored-by: chenxiao <Jaychou1620@Gmail.com>
### What this PR does / why we need it?
Fix pre-create link parameter error, `batch_transfer_sync_write`
requires list.
### Does this PR introduce _any_ user-facing change?
No.
### How was this patch tested?
By CI.
- vLLM version: v0.13.0
- vLLM main:
2f4e6548ef
Signed-off-by: liziyu <liziyu16@huawei.com>
Co-authored-by: liziyu <liziyu16@huawei.com>
### What this PR does / why we need it?
1.replace moe_gating_top_k from torch_npu with custom op
2.enable the renorm function of moe_gating_top_k in softmax scenerio
### Does this PR introduce _any_ user-facing change?
No
### How was this patch tested?
No need test
- vLLM version: v0.13.0
- vLLM main:
7157596103
---------
Signed-off-by: ZCG12345 <2097562023@qq.com>
### What this PR does / why we need it?
1. Accuracy testing no longer compares eager and graph modes; instead,
it directly extracts the golden result under the graph mode
configuration (the implicit purpose of this case is to verify whether
modifications affect existing results)
2. Next step: finer-grained supervision of logits/sampler results
### Does this PR introduce _any_ user-facing change?
### How was this patch tested?
- vLLM version: release/v0.13.0
- vLLM main:
254f6b9867
Signed-off-by: wangli <wangli858794774@gmail.com>
Currently, the vllm pull request
(https://github.com/vllm-project/vllm/pull/24252) is causing operator
fusion to fail. This issue was previously fixed by patching the backend.
The root cause has been identified, and the problem can be resolved with
this pull request.
- vLLM version: v0.13.0
- vLLM main:
2f4e6548ef
---------
Signed-off-by: wxsIcey <1790571317@qq.com>
### What this PR does / why we need it?
Add qwen3-8b nightly test
- vLLM version: v0.13.0
- vLLM main:
7157596103
---------
Signed-off-by: wxsIcey <1790571317@qq.com>
### What this PR does / why we need it?
Import global var form vllm instead of overwirte it, so that we could
use the correct global variant value
- vLLM version: v0.13.0
- vLLM main:
5326c89803
---------
Signed-off-by: MengqingCao <cmq0113@163.com>
## What this PR does / why we need it?
This PR fixes the `AttentionMaskBuilder` singleton initialization issue
introduced in PR #4779 and removes the unused `pcp_prefill_mask` field.
### Background
After PR #4779 made `AttentionMaskBuilder` a singleton with `@singleton`
decorator, the class constructor now requires a `device` parameter.
However, two initialization sites were still using the old parameterless
constructor, causing failures.
### Changes
1. **Fix singleton initialization**
- Fixed `AttentionMaskBuilder()` → `AttentionMaskBuilder(self.device)`
in `AscendMLAMetadataBuilder.__init__()`
- Fixed `AttentionMaskBuilder()` → `AttentionMaskBuilder(self.device)`
in `AscendAttentionMetadataBuilder.__init__()`
2. **Remove unused field**
- Removed `pcp_prefill_mask` field from
`AscendPrefillContextParallelMetadata` (never used in codebase)
- Updated related test assertions
### Related
- Issue #5463
- PR #4779 (Unify all mask generation methods)
- PR #5389 (Make AttentionMaskBuilder singleton)
## Does this PR introduce _any_ user-facing change?
No. This is an internal refactoring.
## How was this patch tested?
- ✅ Local testing: No linter errors
- ✅ Unit tests for attention modules verified
- ⏳ CI pipeline
Signed-off-by: lico67373 <918688502@qq.com>
Co-authored-by: weijinqian0 <1184188277@qq.com>
move image and wheel job to schedule way to save CI resource
- vLLM version: v0.13.0
- vLLM main:
2f4e6548ef
Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
### What this PR does / why we need it?
When launching the service in the scenario where the
cudagraph_mode is set to FULL and Eagle3 acceleration is enabled for
inference, an error in fia will cause graph capture to fail. This PR
fixes the issue.
### Does this PR introduce _any_ user-facing change?
### How was this patch tested?
- vLLM version: v0.13.0
- vLLM main:
7157596103
Signed-off-by: WithHades <244036962@qq.com>
### What this PR does / why we need it?
Resolving the issue of insufficient resources during service operation
when PCP is enabled in a piecewise scenario.
When enabling PCP and executing in piecewise mode, the curl request
fails due to insufficient resources, resulting in the error message "The
resources are insufficient." Through profiling analysis, it was found
that the PCP communication domain also occupies streams and consumes
resources. Therefore, when updating aclgraph sizes, the PCP
communication domain needs to be taken into account.
### Does this PR introduce _any_ user-facing change?
### How was this patch tested?
- vLLM version: release/v0.13.0
- vLLM main:
bc0a5a0c08
---------
Signed-off-by: weiguihua2 <weiguihua2@huawei.com>
1. speed up e2e light test.
2. create `2-cards` and `4-cards` folder in multicard
3. move ops to nightly
4. run test in Alphabetical Order
- vLLM version: v0.13.0
- vLLM main:
8be6432bda
Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
### What this PR does / why we need it?
Add the model to CI for subsequent addition of nightly test cases:
- moonshotai/Kimi-K2-Thinking
- vllm-ascend/Kimi-K2-Instruct-W8A8
### How was this patch tested?
- vLLM version: v0.13.0
- vLLM main:
2f4e6548ef
---------
Signed-off-by: MrZ20 <2609716663@qq.com>
Signed-off-by: wangli <wangli858794774@gmail.com>
Co-authored-by: wangli <wangli858794774@gmail.com>
### What this PR does / why we need it?
To adapt different shapes of the KV cache, UCM optimized the
initialization of store by moving it into `register_kv_caches`.
Therefore, this update adds `register_kv_caches` interface to
UCMConnectorV1.
### How was this patch tested?
- vLLM version: v0.13.0
- vLLM main:
2f4e6548ef
Signed-off-by: UnifiedCacheManager <unifiedcachem@163.com>
#### What this PR does / why we need it?
This PR adapt DispatchGmmCombineDecode operator to eplb tensor list and
expert token numbers.
This operator support gmm1, gmm2, gmm1Scale and gmm2Scale in format of
list.
This operator support couting how many token each local expert recieves
by expertTokensNum .
- vLLM version: v0.13.0
- vLLM main:
7157596103
More info about this operator, please refer to RFC: issue
https://github.com/vllm-project/vllm-ascend/issues/5476
# What this PR does / why we need it?
Add DeepSeek-V3.2-W8A8 dual-node nightly CI test and update A3 nightly
test configuration:
1. Add DeepSeek-V3.2-W8A8 dual-node test:
tests/e2e/nightly/multi_node/config/DeepSeek-V3_2-W8A8-A3-dual-nodes.yaml
- 2 nodes, 16 NPUs per node (32 NPUs total)
- Configuration: 2P+1D (data-parallel-size=4, tensor-parallel-size=8,
data-parallel-size-local=2)
- Includes performance and accuracy benchmarks with GSM8K dataset
2. Update A3 nightly workflow: .github/workflows/nightly_test_a3.yaml
- Added DeepSeek-V3.2-W8A8 dual-node test to the A3 nightly test matrix
- Test name: multi-node-dpsk3.2-2node
3. Improve test scripts: Updated
.github/workflows/_e2e_nightly_multi_node.yaml and related scripts for
better multi-node testing support
test on A3 instances
- Performance baseline: 1 (threshold: 0.97)
- Accuracy baseline: 95% (threshold: 5%)
- Test dataset: GSM8K with 512 prompts for performance, gsm8k-lite for
accuracy
---------
Signed-off-by: guozr <guozr1997@hotmail.com>
Co-authored-by: guozr <guozr1997@hotmail.com>
### What this PR does / why we need it?
Purpose: initialize padded slot mapping buffer to prevent garbage
values.
In PCP mode, the `pcp_padded_slot_mapping` buffer is reused across
invocations. Without explicit initialization, this buffer retain stale
values from previous runs, which can lead to incorrect results.
This change ensures the buffer is filled with -1.
### Does this PR introduce _any_ user-facing change?
No.
### How was this patch tested?
- vLLM version: v0.13.0
- vLLM main:
2f4e6548ef
---------
Signed-off-by: F.Liu <liufeng248@huawei.com>
Co-authored-by: F.Liu <liufeng248@huawei.com>
### What this PR does / why we need it?
1. add `COMPILATION_PASS_KEY` constant
2. clean up useless platform interface `empty_cache`, `synchronize`,
`mem_get_info`, `clear_npu_memory`
3. rename `CUSTOM_OP_REGISTERED` to `_CUSTOM_OP_REGISTERED`
4. remove uesless env `VLLM_ENABLE_CUDAGRAPH_GC`
NPUPlatform is the interface called by vLLM. Do not call it inner
vllm-ascend.
### Does this PR introduce _any_ user-facing change?
This PR is just a cleanup. All CI should pass.
### How was this patch tested?
- vLLM version: v0.13.0
- vLLM main:
7157596103
Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
### What this PR does / why we need it?
This PR builds upon PR
https://github.com/vllm-project/vllm-ascend/pull/5011 and aims to
further enhance the npu_graph_ex_passes module. Based on prior work, we
have added graph optimization support for the add_rms_quant fused
operator in scenarios where a bias term is present—ensuring the fusion
pattern is correctly registered and matched into the computation graph.
For validation, we switched to the Qwen3-235B-A22B-W8A8 model for
SPPatternWithBias and Qwen3-32B model for SPPattern. Benchmark results
show that, compared to the unfused baseline, enabling this fusion pass
significantly improves inference throughput for W8A8 quantized models.
For more details can refer to the
RFC:https://github.com/vllm-project/vllm-ascend/issues/4715
### Does this PR introduce _any_ user-facing change?
no
### How was this patch tested?
```
llm = LLM(
model=model,
tensor_parallel_size=GPUs_per_dp_rank,
enforce_eager=False,
enable_expert_parallel=enable_expert_parallel,
trust_remote_code=trust_remote_code,
gpu_memory_utilization=0.98,
max_num_batched_tokens=512,
# load_format="dummy",
max_model_len=2048,
max_num_seqs=16,
quantization="ascend",
additional_config={
"refresh": True,
"enable_npugraph_ex": True
},
compilation_config={
"cudagraph_capture_sizes": [8, 16],
"cudagraph_mode": "FULL_DECODE_ONLY",
},
)
if profile_dir:
llm.start_profile()
outputs = llm.generate(prompts, sampling_params)
if profile_dir:
llm.stop_profile()
for i, output in enumerate(outputs):
if i >= 5:
break
prompt = output.prompt
generated_text = output.outputs[0].text
print(
f"DP rank {global_dp_rank}, Prompt: {prompt!r}, "
f"Generated text: {generated_text!r}"
)
```
- vLLM version: v0.13.0
- vLLM main:
7157596103
Signed-off-by: cjian <2318164299@qq.com>
### What this PR does / why we need it?
[Bugfix] fix dcp_only bug and add e2e accuracy test for dcp only and pcp
only
this pr fix the bug of accuracy test when decode_parallel_size>1 and
prefill_context_parallel_size=1.
### Does this PR introduce _any_ user-facing change?
NO
### How was this patch tested?
- vLLM version: v0.13.0
- vLLM main:
7157596103
---------
Signed-off-by: zhenwenqi2024 <zhenwenqi_2022@qq.com>
### What this PR does / why we need it?
Revert PR 5253 to fix the smoking problem
### Does this PR introduce _any_ user-facing change?
Does not.
### How was this patch tested?
It was tested in the failure case.
Signed-off-by: Rifa <865071616@qq.com>
### What this PR does / why we need it?
[P/D] Performance enhancement of Layerwise connector in TP asymmetric
scenarios
1. Session fusion: For transmission tasks at each layer, aggregate
transmission tasks with the same destination and merge them into a
single task for assignment.
2. Alltoall aggregation: For TP asymmetric scenarios, perform all
alltoall operations at once according to the block granularity for all
requests.
[RFC]: CDCP Scheduling for Disaggregated Prefilling with KV Cache
Layerwise Push Support
https://github.com/vllm-project/vllm-ascend/issues/4842
### Does this PR introduce _any_ user-facing change?
### How was this patch tested?
- vLLM version: v0.13.0
- vLLM main:
45c1ca1ca1
---------
Signed-off-by: liziyu <liziyu16@huawei.com>
Signed-off-by: nwpu-zxr <zhouxuerong2@huawei.com>
Signed-off-by: wangxiaoteng <wangxiaoteng@huawei.com>
Co-authored-by: nwpu-zxr <zhouxuerong2@huawei.com>
Co-authored-by: wangxiaoteng <wangxiaoteng@huawei.com>
### What this PR does / why we need it?
There was an accuracy issue with **Qwen3-Next-80B-A3B-Instruct-W8A8**
model in the old version of **Triton-Ascend**, so, we are now adding one
nightly test to maintain it.
### Does this PR introduce _any_ user-facing change?
N/A
### How was this patch tested?
- vLLM version: v0.13.0
- vLLM main:
7157596103
Signed-off-by: IncSec <1790766300@qq.com>
### What this PR does / why we need it?
When using the swa parameter in fia, headDim does not currently support
256, and when gemma3's headDim is equal to 256, an error will occur.
Therefore, code rollback is required, and it will be incorporated after
cann supports it.
### Does this PR introduce _any_ user-facing change?
Remove swa parameter of fia.
### How was this patch tested?
- vLLM version: v0.13.0
- vLLM main:
7157596103
---------
Signed-off-by: nsdie <yeyifan@huawei.com>
Co-authored-by: Mengqing Cao <cmq0113@163.com>
#### Overview
This PR fixes a shape mismatch bug between `expert_placement_map` and
`log2phy_expert_map` when **redundant experts** are enabled in the
vLLM-Ascend platform. The issue occurred during the initialization of
expert maps and their updates via EPLB (Expert Load Balancer)
adjustment, leading to potential tensor shape errors and incorrect
expert routing in distributed MoE deployments.
#### Key Changes
1. **Unify expert map shape calculation logic**
- Ensure the shape of `expert_placement_map` and `log2phy_expert_map`
strictly aligns with the total number of experts (including redundant
experts) during initialization.
- Update the shape adjustment logic in EPLB dynamic update process to
match the initial expert map dimensions.
2. **Add shape consistency checks**
- Add assertion statements to verify the shape consistency of the two
maps after initialization and EPLB adjustment, preventing silent shape
mismatches in subsequent operations.
#### Impact
- Resolves tensor shape errors when using redundant experts with EPLB on
Ascend platform.
- Ensures correct expert routing and load balancing for MoE models with
redundant expert configurations.
- No breaking changes to existing functionality; compatible with
non-redundant expert deployments.
- vLLM version: release/v0.13.0
- vLLM main:
ad32e3e19c
---------
Signed-off-by: Che Ruan <cr623@ic.ac.uk>
Signed-off-by: shenchuxiaofugui <1311027364@qq.com>
Co-authored-by: Che Ruan <cr623@ic.ac.uk>
Co-authored-by: shenchuxiaofugui <1311027364@qq.com>
### What this PR does / why we need it?
This PR aims to delete redundant methods in mtp_proposer. All the
deleted methods now can be found in eagle_proposer. We also remove some
methods in eagle_proposer since they are identical to those in
vllm-eagle.
### Does this PR introduce _any_ user-facing change?
N/A
### How was this patch tested?
by ci
- vLLM version: release/v0.13.0
- vLLM main:
81786c8774
---------
Signed-off-by: Zetong Li <slippersss@126.com>
### What this PR does / why we need it?
Pin the xlite version to avoid CI failures during its upgrade.
- vLLM version: v0.13.0
- vLLM main:
7157596103
Signed-off-by: changdawei1 <changdawei3@huawei.com>
### What this PR does / why we need it?
Upgrade vllm commit to 0106
- vLLM version: v0.13.0
- vLLM main:
8be6432bda
Signed-off-by: wjunLu <wjunlu217@gmail.com>
### What this PR does / why we need it?
update bisheng version in 20260105
- vLLM version: v0.13.0
- vLLM main:
8be6432bda
Signed-off-by: Meihan-chen <jcccx.cmh@gmail.com>
## Purpose
This PR add unit test for `compute_slot_mapping` function in
`block_table.py` with various `pcp_size` & `dcp_size` &
`cp_kv_cache_interleave_size`.
## Test Plan
```
pytest tests/ut/worker/test_block_table.py
```
## Test Result
```
==== 3 passed, 2 warnings in 0.20s ====
```
- vLLM version: v0.12.0
- vLLM main:
ad32e3e19c
---------
Signed-off-by: QiuChunshuo <qiuchunshuo@huawei.com>
### What this PR does / why we need it?
mv ops to correct path
:`tests/e2e/nightly/single_node/ops/singlecard_ops/triton`
Signed-off-by: wangli <wangli858794774@gmail.com>
### What this PR does / why we need it?
Fix Smoke Testing Bug for DSR1 longseq
We need to make this change because the daily smoke test case is
throwing an error: "max_tokens or max_completion_tokens is too large:
32768.This model's maximum context length is 32768 tokens and your
request has 128 input tokens". We encounter this error due to
max-out-len equals to max-model-len. We can fix this error by increasing
max-model-len argument in the script.
### Does this PR introduce _any_ user-facing change?
### How was this patch tested?
- vLLM version: v0.13.0
- vLLM main:
7157596103
Signed-off-by: daishixun <dsxsteven@sina.com>
